GaetanMichelet
commited on
Commit
•
e550a22
1
Parent(s):
b0077af
Model save
Browse files- README.md +75 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +468 -0
README.md
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.3
|
3 |
+
library_name: peft
|
4 |
+
license: apache-2.0
|
5 |
+
tags:
|
6 |
+
- trl
|
7 |
+
- sft
|
8 |
+
- generated_from_trainer
|
9 |
+
model-index:
|
10 |
+
- name: Mistral-7B_task-1_180-samples_config-2_auto
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# Mistral-7B_task-1_180-samples_config-2_auto
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the None dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 1.3901
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0001
|
41 |
+
- train_batch_size: 1
|
42 |
+
- eval_batch_size: 1
|
43 |
+
- seed: 42
|
44 |
+
- distributed_type: multi-GPU
|
45 |
+
- gradient_accumulation_steps: 16
|
46 |
+
- total_train_batch_size: 16
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: cosine
|
49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
50 |
+
- num_epochs: 50
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
56 |
+
| 1.0549 | 0.9412 | 8 | 0.9900 |
|
57 |
+
| 0.8222 | 2.0 | 17 | 0.8564 |
|
58 |
+
| 0.7511 | 2.9412 | 25 | 0.7989 |
|
59 |
+
| 0.5644 | 4.0 | 34 | 0.7933 |
|
60 |
+
| 0.3911 | 4.9412 | 42 | 0.9403 |
|
61 |
+
| 0.2483 | 6.0 | 51 | 1.0407 |
|
62 |
+
| 0.1183 | 6.9412 | 59 | 1.1922 |
|
63 |
+
| 0.0848 | 8.0 | 68 | 1.2871 |
|
64 |
+
| 0.0657 | 8.9412 | 76 | 1.3606 |
|
65 |
+
| 0.0445 | 10.0 | 85 | 1.3682 |
|
66 |
+
| 0.0398 | 10.9412 | 93 | 1.3901 |
|
67 |
+
|
68 |
+
|
69 |
+
### Framework versions
|
70 |
+
|
71 |
+
- PEFT 0.12.0
|
72 |
+
- Transformers 4.44.0
|
73 |
+
- Pytorch 2.1.2+cu121
|
74 |
+
- Datasets 2.20.0
|
75 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 167832240
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:880845919083852056f38bb89a765a25120b554fda03861744d562a315c5eb9c
|
3 |
size 167832240
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.941176470588236,
|
3 |
+
"total_flos": 2.518477265751245e+16,
|
4 |
+
"train_loss": 0.3923261252141768,
|
5 |
+
"train_runtime": 1825.5377,
|
6 |
+
"train_samples": 136,
|
7 |
+
"train_samples_per_second": 3.725,
|
8 |
+
"train_steps_per_second": 0.219
|
9 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.941176470588236,
|
3 |
+
"total_flos": 2.518477265751245e+16,
|
4 |
+
"train_loss": 0.3923261252141768,
|
5 |
+
"train_runtime": 1825.5377,
|
6 |
+
"train_samples": 136,
|
7 |
+
"train_samples_per_second": 3.725,
|
8 |
+
"train_steps_per_second": 0.219
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,468 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7932864427566528,
|
3 |
+
"best_model_checkpoint": "data/Mistral-7B_task-1_180-samples_config-2_auto/checkpoint-34",
|
4 |
+
"epoch": 10.941176470588236,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 93,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.11764705882352941,
|
13 |
+
"grad_norm": 2.450850009918213,
|
14 |
+
"learning_rate": 2.5e-06,
|
15 |
+
"loss": 1.0532,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.23529411764705882,
|
20 |
+
"grad_norm": 2.664348602294922,
|
21 |
+
"learning_rate": 5e-06,
|
22 |
+
"loss": 1.038,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.47058823529411764,
|
27 |
+
"grad_norm": 2.440438985824585,
|
28 |
+
"learning_rate": 1e-05,
|
29 |
+
"loss": 1.1264,
|
30 |
+
"step": 4
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.7058823529411765,
|
34 |
+
"grad_norm": 2.530811309814453,
|
35 |
+
"learning_rate": 1.5e-05,
|
36 |
+
"loss": 1.0888,
|
37 |
+
"step": 6
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.9411764705882353,
|
41 |
+
"grad_norm": 2.1098403930664062,
|
42 |
+
"learning_rate": 2e-05,
|
43 |
+
"loss": 1.0549,
|
44 |
+
"step": 8
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.9411764705882353,
|
48 |
+
"eval_loss": 0.9900373816490173,
|
49 |
+
"eval_runtime": 14.9949,
|
50 |
+
"eval_samples_per_second": 2.401,
|
51 |
+
"eval_steps_per_second": 2.401,
|
52 |
+
"step": 8
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.1764705882352942,
|
56 |
+
"grad_norm": 1.6433088779449463,
|
57 |
+
"learning_rate": 2.5e-05,
|
58 |
+
"loss": 0.9855,
|
59 |
+
"step": 10
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.4117647058823528,
|
63 |
+
"grad_norm": 1.2828103303909302,
|
64 |
+
"learning_rate": 3e-05,
|
65 |
+
"loss": 0.8472,
|
66 |
+
"step": 12
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.6470588235294117,
|
70 |
+
"grad_norm": 1.1351650953292847,
|
71 |
+
"learning_rate": 3.5e-05,
|
72 |
+
"loss": 0.8693,
|
73 |
+
"step": 14
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 1.8823529411764706,
|
77 |
+
"grad_norm": 1.176932692527771,
|
78 |
+
"learning_rate": 4e-05,
|
79 |
+
"loss": 0.8222,
|
80 |
+
"step": 16
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 2.0,
|
84 |
+
"eval_loss": 0.8563865423202515,
|
85 |
+
"eval_runtime": 14.9839,
|
86 |
+
"eval_samples_per_second": 2.403,
|
87 |
+
"eval_steps_per_second": 2.403,
|
88 |
+
"step": 17
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 2.1176470588235294,
|
92 |
+
"grad_norm": 1.0588655471801758,
|
93 |
+
"learning_rate": 4.5e-05,
|
94 |
+
"loss": 0.7559,
|
95 |
+
"step": 18
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 2.3529411764705883,
|
99 |
+
"grad_norm": 1.0662921667099,
|
100 |
+
"learning_rate": 5e-05,
|
101 |
+
"loss": 0.7459,
|
102 |
+
"step": 20
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 2.588235294117647,
|
106 |
+
"grad_norm": 0.9237774610519409,
|
107 |
+
"learning_rate": 5.500000000000001e-05,
|
108 |
+
"loss": 0.7367,
|
109 |
+
"step": 22
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 2.8235294117647056,
|
113 |
+
"grad_norm": 0.9688161015510559,
|
114 |
+
"learning_rate": 6e-05,
|
115 |
+
"loss": 0.7511,
|
116 |
+
"step": 24
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 2.9411764705882355,
|
120 |
+
"eval_loss": 0.798934280872345,
|
121 |
+
"eval_runtime": 14.9861,
|
122 |
+
"eval_samples_per_second": 2.402,
|
123 |
+
"eval_steps_per_second": 2.402,
|
124 |
+
"step": 25
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 3.0588235294117645,
|
128 |
+
"grad_norm": 0.9950412511825562,
|
129 |
+
"learning_rate": 6.500000000000001e-05,
|
130 |
+
"loss": 0.6961,
|
131 |
+
"step": 26
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 3.2941176470588234,
|
135 |
+
"grad_norm": 0.9355748295783997,
|
136 |
+
"learning_rate": 7e-05,
|
137 |
+
"loss": 0.6115,
|
138 |
+
"step": 28
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 3.5294117647058822,
|
142 |
+
"grad_norm": 1.1376206874847412,
|
143 |
+
"learning_rate": 7.500000000000001e-05,
|
144 |
+
"loss": 0.6316,
|
145 |
+
"step": 30
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 3.764705882352941,
|
149 |
+
"grad_norm": 1.149610996246338,
|
150 |
+
"learning_rate": 8e-05,
|
151 |
+
"loss": 0.595,
|
152 |
+
"step": 32
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 4.0,
|
156 |
+
"grad_norm": 1.1590631008148193,
|
157 |
+
"learning_rate": 8.5e-05,
|
158 |
+
"loss": 0.5644,
|
159 |
+
"step": 34
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 4.0,
|
163 |
+
"eval_loss": 0.7932864427566528,
|
164 |
+
"eval_runtime": 15.0042,
|
165 |
+
"eval_samples_per_second": 2.399,
|
166 |
+
"eval_steps_per_second": 2.399,
|
167 |
+
"step": 34
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 4.235294117647059,
|
171 |
+
"grad_norm": 1.2311124801635742,
|
172 |
+
"learning_rate": 9e-05,
|
173 |
+
"loss": 0.4864,
|
174 |
+
"step": 36
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 4.470588235294118,
|
178 |
+
"grad_norm": 1.1586334705352783,
|
179 |
+
"learning_rate": 9.5e-05,
|
180 |
+
"loss": 0.4053,
|
181 |
+
"step": 38
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 4.705882352941177,
|
185 |
+
"grad_norm": 1.3652201890945435,
|
186 |
+
"learning_rate": 0.0001,
|
187 |
+
"loss": 0.4336,
|
188 |
+
"step": 40
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 4.9411764705882355,
|
192 |
+
"grad_norm": 1.6470670700073242,
|
193 |
+
"learning_rate": 9.999238475781957e-05,
|
194 |
+
"loss": 0.3911,
|
195 |
+
"step": 42
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 4.9411764705882355,
|
199 |
+
"eval_loss": 0.9402649998664856,
|
200 |
+
"eval_runtime": 14.9808,
|
201 |
+
"eval_samples_per_second": 2.403,
|
202 |
+
"eval_steps_per_second": 2.403,
|
203 |
+
"step": 42
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 5.176470588235294,
|
207 |
+
"grad_norm": 1.2947176694869995,
|
208 |
+
"learning_rate": 9.99695413509548e-05,
|
209 |
+
"loss": 0.2718,
|
210 |
+
"step": 44
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 5.411764705882353,
|
214 |
+
"grad_norm": 1.3543827533721924,
|
215 |
+
"learning_rate": 9.99314767377287e-05,
|
216 |
+
"loss": 0.2345,
|
217 |
+
"step": 46
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"epoch": 5.647058823529412,
|
221 |
+
"grad_norm": 1.7274010181427002,
|
222 |
+
"learning_rate": 9.987820251299122e-05,
|
223 |
+
"loss": 0.2069,
|
224 |
+
"step": 48
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 5.882352941176471,
|
228 |
+
"grad_norm": 1.9732545614242554,
|
229 |
+
"learning_rate": 9.980973490458728e-05,
|
230 |
+
"loss": 0.2483,
|
231 |
+
"step": 50
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 6.0,
|
235 |
+
"eval_loss": 1.0406984090805054,
|
236 |
+
"eval_runtime": 14.9922,
|
237 |
+
"eval_samples_per_second": 2.401,
|
238 |
+
"eval_steps_per_second": 2.401,
|
239 |
+
"step": 51
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 6.117647058823529,
|
243 |
+
"grad_norm": 1.330666422843933,
|
244 |
+
"learning_rate": 9.972609476841367e-05,
|
245 |
+
"loss": 0.187,
|
246 |
+
"step": 52
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 6.352941176470588,
|
250 |
+
"grad_norm": 1.65781831741333,
|
251 |
+
"learning_rate": 9.962730758206611e-05,
|
252 |
+
"loss": 0.1408,
|
253 |
+
"step": 54
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 6.588235294117647,
|
257 |
+
"grad_norm": 1.3344322443008423,
|
258 |
+
"learning_rate": 9.951340343707852e-05,
|
259 |
+
"loss": 0.1191,
|
260 |
+
"step": 56
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 6.823529411764706,
|
264 |
+
"grad_norm": 1.7343159914016724,
|
265 |
+
"learning_rate": 9.938441702975689e-05,
|
266 |
+
"loss": 0.1183,
|
267 |
+
"step": 58
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 6.9411764705882355,
|
271 |
+
"eval_loss": 1.192215085029602,
|
272 |
+
"eval_runtime": 14.9851,
|
273 |
+
"eval_samples_per_second": 2.402,
|
274 |
+
"eval_steps_per_second": 2.402,
|
275 |
+
"step": 59
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 7.0588235294117645,
|
279 |
+
"grad_norm": 1.408785104751587,
|
280 |
+
"learning_rate": 9.924038765061042e-05,
|
281 |
+
"loss": 0.1173,
|
282 |
+
"step": 60
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 7.294117647058823,
|
286 |
+
"grad_norm": 1.2464698553085327,
|
287 |
+
"learning_rate": 9.908135917238321e-05,
|
288 |
+
"loss": 0.0852,
|
289 |
+
"step": 62
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 7.529411764705882,
|
293 |
+
"grad_norm": 1.4764487743377686,
|
294 |
+
"learning_rate": 9.890738003669029e-05,
|
295 |
+
"loss": 0.0745,
|
296 |
+
"step": 64
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 7.764705882352941,
|
300 |
+
"grad_norm": 2.1253485679626465,
|
301 |
+
"learning_rate": 9.871850323926177e-05,
|
302 |
+
"loss": 0.088,
|
303 |
+
"step": 66
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 8.0,
|
307 |
+
"grad_norm": 1.791509985923767,
|
308 |
+
"learning_rate": 9.851478631379982e-05,
|
309 |
+
"loss": 0.0848,
|
310 |
+
"step": 68
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 8.0,
|
314 |
+
"eval_loss": 1.2871198654174805,
|
315 |
+
"eval_runtime": 14.983,
|
316 |
+
"eval_samples_per_second": 2.403,
|
317 |
+
"eval_steps_per_second": 2.403,
|
318 |
+
"step": 68
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 8.235294117647058,
|
322 |
+
"grad_norm": 1.1020724773406982,
|
323 |
+
"learning_rate": 9.829629131445342e-05,
|
324 |
+
"loss": 0.0543,
|
325 |
+
"step": 70
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 8.470588235294118,
|
329 |
+
"grad_norm": 1.1706922054290771,
|
330 |
+
"learning_rate": 9.806308479691595e-05,
|
331 |
+
"loss": 0.0547,
|
332 |
+
"step": 72
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 8.705882352941176,
|
336 |
+
"grad_norm": 1.9083949327468872,
|
337 |
+
"learning_rate": 9.781523779815179e-05,
|
338 |
+
"loss": 0.0682,
|
339 |
+
"step": 74
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 8.941176470588236,
|
343 |
+
"grad_norm": 2.0456669330596924,
|
344 |
+
"learning_rate": 9.755282581475769e-05,
|
345 |
+
"loss": 0.0657,
|
346 |
+
"step": 76
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 8.941176470588236,
|
350 |
+
"eval_loss": 1.3606077432632446,
|
351 |
+
"eval_runtime": 14.982,
|
352 |
+
"eval_samples_per_second": 2.403,
|
353 |
+
"eval_steps_per_second": 2.403,
|
354 |
+
"step": 76
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 9.176470588235293,
|
358 |
+
"grad_norm": 1.3962442874908447,
|
359 |
+
"learning_rate": 9.727592877996585e-05,
|
360 |
+
"loss": 0.0433,
|
361 |
+
"step": 78
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 9.411764705882353,
|
365 |
+
"grad_norm": 1.3246513605117798,
|
366 |
+
"learning_rate": 9.698463103929542e-05,
|
367 |
+
"loss": 0.0416,
|
368 |
+
"step": 80
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 9.647058823529411,
|
372 |
+
"grad_norm": 1.652782917022705,
|
373 |
+
"learning_rate": 9.667902132486009e-05,
|
374 |
+
"loss": 0.0564,
|
375 |
+
"step": 82
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 9.882352941176471,
|
379 |
+
"grad_norm": 1.3555011749267578,
|
380 |
+
"learning_rate": 9.635919272833938e-05,
|
381 |
+
"loss": 0.0445,
|
382 |
+
"step": 84
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 10.0,
|
386 |
+
"eval_loss": 1.36819589138031,
|
387 |
+
"eval_runtime": 14.9982,
|
388 |
+
"eval_samples_per_second": 2.4,
|
389 |
+
"eval_steps_per_second": 2.4,
|
390 |
+
"step": 85
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 10.117647058823529,
|
394 |
+
"grad_norm": 0.7261105179786682,
|
395 |
+
"learning_rate": 9.602524267262203e-05,
|
396 |
+
"loss": 0.0392,
|
397 |
+
"step": 86
|
398 |
+
},
|
399 |
+
{
|
400 |
+
"epoch": 10.352941176470589,
|
401 |
+
"grad_norm": 1.1909233331680298,
|
402 |
+
"learning_rate": 9.567727288213005e-05,
|
403 |
+
"loss": 0.0336,
|
404 |
+
"step": 88
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 10.588235294117647,
|
408 |
+
"grad_norm": 1.4687535762786865,
|
409 |
+
"learning_rate": 9.53153893518325e-05,
|
410 |
+
"loss": 0.0445,
|
411 |
+
"step": 90
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 10.823529411764707,
|
415 |
+
"grad_norm": 1.1714824438095093,
|
416 |
+
"learning_rate": 9.493970231495835e-05,
|
417 |
+
"loss": 0.0398,
|
418 |
+
"step": 92
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 10.941176470588236,
|
422 |
+
"eval_loss": 1.3901447057724,
|
423 |
+
"eval_runtime": 14.9942,
|
424 |
+
"eval_samples_per_second": 2.401,
|
425 |
+
"eval_steps_per_second": 2.401,
|
426 |
+
"step": 93
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 10.941176470588236,
|
430 |
+
"step": 93,
|
431 |
+
"total_flos": 2.518477265751245e+16,
|
432 |
+
"train_loss": 0.3923261252141768,
|
433 |
+
"train_runtime": 1825.5377,
|
434 |
+
"train_samples_per_second": 3.725,
|
435 |
+
"train_steps_per_second": 0.219
|
436 |
+
}
|
437 |
+
],
|
438 |
+
"logging_steps": 2,
|
439 |
+
"max_steps": 400,
|
440 |
+
"num_input_tokens_seen": 0,
|
441 |
+
"num_train_epochs": 50,
|
442 |
+
"save_steps": 25,
|
443 |
+
"stateful_callbacks": {
|
444 |
+
"EarlyStoppingCallback": {
|
445 |
+
"args": {
|
446 |
+
"early_stopping_patience": 7,
|
447 |
+
"early_stopping_threshold": 0.0
|
448 |
+
},
|
449 |
+
"attributes": {
|
450 |
+
"early_stopping_patience_counter": 0
|
451 |
+
}
|
452 |
+
},
|
453 |
+
"TrainerControl": {
|
454 |
+
"args": {
|
455 |
+
"should_epoch_stop": false,
|
456 |
+
"should_evaluate": false,
|
457 |
+
"should_log": false,
|
458 |
+
"should_save": true,
|
459 |
+
"should_training_stop": true
|
460 |
+
},
|
461 |
+
"attributes": {}
|
462 |
+
}
|
463 |
+
},
|
464 |
+
"total_flos": 2.518477265751245e+16,
|
465 |
+
"train_batch_size": 1,
|
466 |
+
"trial_name": null,
|
467 |
+
"trial_params": null
|
468 |
+
}
|